Research map / 3D vision / Face and shape
DreamWaltz: Make a Scene with Complex 3D Animatable Avatars
Yukun Huang, Jianan Wang, Ailing Zeng, He CAO, Xianbiao Qi, Yukai Shi, Zheng-Jun Zha, Lei Zhang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add occlusion culling of invisible facial keypoints in skeleton conditioning (ablation) | DreamWaltz · ControlNet back-view generation and text-to-3D avatars | multi-face (Janus) artifacts, qualitative | Improved |
| use learned animation with density weighting and pose prior instead of Inverse-LBS or AvatarCraft (ablation) | DreamWaltz · avatar animation | animation quality, qualitative | Improved |
| two-stage training (canonical then animatable) instead of single-stage (ablation) | DreamWaltz · dance motion sequence animation | visual quality, qualitative | Improved |
| add scene-level fine-tuning with 3D-consistent SDS (ablation) | composed avatar scenes · scene composition | visual quality, qualitative | Improved |
| skeleton conditioning instead of depth conditioning for SDS (ablation) | DreamWaltz · superman prompt SDS gradients | supervision flexibility, qualitative | Improved |
| Stage II random-pose optimization on canonical avatar (ablation) | DreamWaltz · canonical avatar appearance | visual detail, qualitative | Mixed |
| DreamWaltz vs DreamAvatar (best competitor) in user preference | DreamWaltz · user study geometry | score (1-4): +0.67 | Improved |
| DreamWaltz vs DreamAvatar (best competitor) in user preference | DreamWaltz · user study texture | score (1-4): +1.01 | Improved |
Later papers that compared with this method
5 later accepted papers report DreamWaltz as a baseline in their results tables; 5 of them report a better number on most of the results they share with it.
- HandDreamer: Zero-Shot Text to 3D Hand Model Generation using Corrective Hand Shape Guidance (CVPR 2026 · better on 3 of 3 shared results)
- DAGSM: Disentangled Avatar Generation with GS-enhanced Mesh (CVPR 2025 · better on 5 of 5 shared results)
- VTON 360: High-Fidelity Virtual Try-On from Any Viewing Direction (CVPR 2025 · better on 1 of 1 shared result)
- AvatarStudio: High-Fidelity and Animatable 3D Avatar Creation from Text (IJCV 2025 · better on 1 of 1 shared result)
- Text to Layer-wise 3D Clothed Human Generation (ECCV 2024 · better on 1 of 1 shared result)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- DreamAvatar: Text-and-Shape Guided 3D Human Avatar Generation via Diffusion Models (CVPR 2024 · similarity 0.90)
- AvatarVerse: High-Quality & Stable 3D Avatar Creation from Text and Pose (AAAI 2024 · similarity 0.89)
- X-Oscar: A Progressive Framework for High-quality Text-guided 3D Animatable Avatar Generation (ICML 2024 · similarity 0.88)
- Make-It-Animatable: An Efficient Framework for Authoring Animation-Ready 3D Characters (CVPR 2025 · similarity 0.87)
- Morphable Diffusion: 3D-Consistent Diffusion for Single-image Avatar Creation (CVPR 2024 · similarity 0.87)
- 3D²-Actor: Learning Pose-Conditioned 3D-Aware Denoiser for Realistic Gaussian Avatar Modeling (AAAI 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
